U.S. patent application number 10/361441 was filed with the patent office on 2003-10-23 for method and apparatus for guiding movement of a freely roaming animal through brain stimulation.
Invention is credited to Chapin, John K., Hawley, Emerson S., Talwar, Sanjiv K., Xu, Shaohua.
Application Number | 20030199944 10/361441 |
Document ID | / |
Family ID | 27734458 |
Filed Date | 2003-10-23 |
United States Patent
Application |
20030199944 |
Kind Code |
A1 |
Chapin, John K. ; et
al. |
October 23, 2003 |
Method and apparatus for guiding movement of a freely roaming
animal through brain stimulation
Abstract
Movement of a freely roaming animal (190), such as a rat, is
guided using electric stimulation of the animal's brain. Cues are
provided to the animal to move forward by stimulating a reward
center of the brain. Cues are provided to the animal to change its
direction by stimulating portions of the animal's brain that
control right and left movements, such as a cortical representation
of whiskers of the animal. Multi-channel, remotely controlled
equipment (140, 145, 150, 350) may be carried by the animal to
enable independent energizing of electrodes attached to different
regions of the animal's brain. A transmitter carried by the animal
may report back data to allow monitoring. A component may be
carried by the animal for carrying out a mission, such as for
search and rescue or surveillance. Groups of animals may be
controlled in real-time by coordinating their movements and
tracking their locations.
Inventors: |
Chapin, John K.; (Atlantic
Beach, NY) ; Talwar, Sanjiv K.; (Brooklyn, NY)
; Xu, Shaohua; (Brooklyn, NY) ; Hawley, Emerson
S.; (Brooklyn, NY) |
Correspondence
Address: |
SCULLY SCOTT MURPHY & PRESSER, PC
400 GARDEN CITY PLAZA
GARDEN CITY
NY
11530
|
Family ID: |
27734458 |
Appl. No.: |
10/361441 |
Filed: |
February 10, 2003 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
60355050 |
Feb 8, 2002 |
|
|
|
Current U.S.
Class: |
607/48 |
Current CPC
Class: |
G06N 3/061 20130101;
A01K 15/021 20130101; A01K 1/031 20130101; A61N 1/0529 20130101;
A61N 1/36017 20130101 |
Class at
Publication: |
607/48 |
International
Class: |
A61N 001/18 |
Goverment Interests
[0002] The Government may have rights in the invention.
Claims
What is claimed is:
1. A method for guiding movement of a freely roaming animal,
comprising: providing cues to the animal to move forward by
stimulating a reward center of the animal's brain; and providing
cues to the animal to change its direction of movement by
stimulating portions of the animal's brain which control left and
right movements.
2. The method of claim 1, wherein: the portions of the animal's
brain which control left and right movements comprise cortical
representations of left or right whiskers of the animal.
3. The method of claim 2, wherein: the cortical representation
comprises somatosensory cortices of the animal's brain.
4. The method of claim 1, wherein: the reward center comprises at
least one of a median forebrain bundle and a ventral tegmental area
of the animal's brain.
5. The method of claim 1, wherein: the reward center and portions
of the animal's brain which control left and right movements are
stimulated by energizing electrodes implanted in the animal's
brain.
6. The method of claim 1, wherein: the reward center and portions
of the animal's brain which control left and right movements are
stimulated using biphasic stimulus pulses.
7. An apparatus for guiding movement of a freely roaming animal,
comprising: electrodes implanted in a reward center of the animal's
brain, and in portions of the animal's brain which control left and
right movements; and means for energizing the electrodes to provide
cues to the animal to move forward by stimulating the reward
center, and to provide cues to the animal to change its direction
of movement by stimulating the portions of the animal's brain which
control left and right movements.
8. The apparatus of claim 7, wherein: the energizing means are
adapted to be carried by the animal.
9. The apparatus of claim 7, wherein: the energizing means are
provided, at least in part, in a backpack that is adapted to be
carried by the animal.
10. The apparatus of claim 7, further comprising: a remotely
controlled receiver adapted to be carried by the animal for
receiving signals for controlling the energizing means.
11. The apparatus of claim 10, wherein: the remotely controlled
receiver is responsive to control signals from a transmitter.
12. The apparatus of claim 7, further comprising: a component
adapted to be carried by the animal for carrying out a mission.
13. The apparatus of claim 7, further comprising: a transmitter
adapted to be carried by the animal for transmitting data.
14. The apparatus of claim 7, wherein: the portions of the animal's
brain which control left and right movements comprise cortical
representations of left and right whiskers of the animal.
15. The apparatus of claim 14, wherein: the cortical representation
comprises somatosensory cortices of the animal's brain.
16. The apparatus of claim 7, wherein: the reward center comprises
at least one of a median forebrain bundle and a ventral tegmental
area of the animal's brain.
17. The apparatus of claim 7, wherein: the reward center and the
portions of the animal's brain which control left and right
movements are stimulated using biphasic stimulus pulses.
18. A method for guiding movement of a plurality of respective
freely roaming animals, comprising: providing cues to each
respective animal to move forward by stimulating a reward center of
the respective animal's brain; and providing cues to each
respective animal to change its direction of movement by
stimulating portions of each respective animal's brain which
control left and right movements of the respective animal.
19. The method of claim 18, wherein: the cues are provided to the
animals to coordinate their movements.
20. An apparatus for guiding movement of a freely roaming animal,
comprising: a remotely controlled receiver adapted to be carried by
the animal; and energizing means responsive to the receiver for
energizing electrodes implanted in different sites in the animal's
brain to provide cues to the animal to move forward and to change
its direction of movement.
21. The apparatus of claim 20, wherein: the receiver and energizing
means enable independent control of pairs of the electrodes.
22. The apparatus of claim 20, wherein: the receiver receives a
multi-signal, wherein at least two of the channels of the signal
enable independent control of at least two associated pairs of the
electrodes.
23. The apparatus of claim 20, wherein: the sites in which the
electrodes are implanted include a reward center and a cortical
representation of whiskers of the animal's brain.
24. The apparatus of claim 23, wherein: the reward center and
cortical representation of whiskers are stimulated using biphasic
stimulus pulses.
25. The apparatus of claim 20, wherein: the energizing means
comprises a microprocessor.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application claims the benefit of U.S. Provisional
Patent Application Serial No. 60/355,050, filed Feb. 8, 2002, and
incorporated herein by reference.
BACKGROUND OF THE INVENTION
[0003] 1. Field of Invention
[0004] The invention relates generally to the field of guiding the
movement of animals and, more specifically, to a method and
apparatus for guiding the movement of a freely roaming animal using
electric stimulation of the animal's brain.
[0005] 2. Description of Related Art
[0006] Humans have long sought to understand how the brain
functions. Examination and testing of laboratory animals has been
undertaken in this regard. Procedures used to train laboratory
animals often incorporate operant learning paradigms in which the
animals are taught to make particular responses to external cues
(e.g., tones) in order to obtain rewards (e.g., food). Moreover,
electrical stimulation in the central nervous system has long been
a tool in neurophysiology. However, previous approaches were
constrained to using electric cables to connect brain-implanted
electrodes to an external stimulator. While in anesthetized animals
cable connections are generally adequate, in wakeful animals (such
as monkeys) they not only limit the subject's freedom of movement,
but also may distract its attention or produce emotional distress.
Limitations in cable length also confine the animal's movement to
small 2-D spaces.
[0007] Accordingly, there is a need for a method and apparatus that
enriches the scope of investigable behavioral paradigms and enables
brain stimulation experiments in animals moving freely in large and
complex 3-dimensional (3D) environments. Moreover, it would be
desirable to be able to use such remotely guided animals for search
and rescue, law enforcement, military and other purposes. The
present invention addresses the above and other issues.
BRIEF SUMMARY OF THE INVENTION
[0008] The present invention describes a system and method for
controlling the movement of a freely roaming animal using electric
stimulation of the animal's brain.
[0009] In one aspect, the invention provides a miniaturized
multi-channel digital tele-stimulation system that allows remote
delivery of stimulations to multiple brain sites of an animal. New
behavioral models can be developed based solely on brain
stimulation for studying the neural correlates of spatial learning.
Moreover, brain stimulation can be used to generate cues and
rewards, where the rewards can act as cues as well, and
reinforcement contingencies can be arranged so that a human
operator or computer can accurately guide the animal remotely, over
arbitrarily defined routes and over varied 3-D terrains. The system
may be built inexpensively using a commercially available radio
modem and microprocessor components.
[0010] In a particular aspect of the invention, a method for
guiding movement of a freely roaming animal includes providing cues
to the animal to move forward by stimulating a reward center of the
animal's brain, and providing cues to the animal to change its
direction of movement by stimulating portions of the animal's brain
which control left and right movements. A corresponding apparatus
is also presented.
[0011] In another aspect, a method for guiding movement of a number
of respective freely roaming animals includes providing cues to
each animal to move forward by stimulating a reward center of the
respective animal's brain, and providing cues to each animal to
change its direction of movement by stimulating portions of the
animal's brain which control left and right movements.
[0012] In another aspect, an apparatus for guiding movement of a
freely roaming animal includes a remotely controlled receiver
adapted to be carried by the animal, and energizing means
responsive to the receiver for energizing electrodes implanted in
different sites in the animal's brain to provide cues to the animal
to move forward and to change its direction of movement. The
receiver may be a multi-channel receiver that independently
controls pairs of electrodes in response to user commands.
BRIEF DESCRIPTION OF THE DRAWINGS
[0013] These and other features, benefits and advantages of the
present invention will become apparent by reference to the
following text and figures, with like reference numbers referring
to like structures across the views, wherein:
[0014] FIG. 1 illustrates a system for stimulating an animal's
brain by remote control;
[0015] FIG. 2(a) illustrates movement of an animal over a
two-dimensional course;
[0016] FIG. 2(b) illustrates movement of an animal over a
three-dimensional course;
[0017] FIG. 3 illustrates a schematic diagram of circuits for a
base station transmitter, and for a receiver carried by an
animal;
[0018] FIG. 4 illustrates a stimulation command string;
[0019] FIG. 5 illustrates a plot showing a number of lever presses
by an animal when applying a constant voltage and a constant
current source; and
[0020] FIG. 6 illustrates a plot showing a number of lever presses
per minute by an animal when applying a constant voltage with
different pulse durations.
DETAILED DESCRIPTION OF THE INVENTION
[0021] By removing physical constraints associated with the
delivery of cues and rewards, learning paradigms based on brain
microstimulation can enable conditioning approaches that help
transcend traditional boundaries in animal learning. Our
experiments applied this paradigm to develop a behavioral model in
which an experimenter is able to guide distant animals in the
manner of intelligent robots.
[0022] FIG. 1 illustrates an overview of a multichannnel
telestimulation system showing the main components of the system
and the signal flow. In one possible approach, a laptop personal
computer 100 receives commands from an operator, e.g., via specific
keystrokes, for guiding movement of a freely roaming animal 190,
such as a rat. The laptop 100 sends a control signal to a base
station 110 via a serial RS232 port 115. An optocoupler 120
processes the signal and provides it to a transmitter 125 as a
transistor-transistor logic (TTL) signal. The transmitter 125
transmits the signal via antenna 130 and a radio link to an antenna
140 of a receiver 145, which is carried by the animal 190, such as
in a backpack 160 which is secured to the animal using a harness
165 (available from Harvard Apparatus, Holliston, Mass.). The
backpack 160 measured 48 mm.times.23 mm.times.19 mm and weighed 28
Gms, and is worn by the rat 190 by means of mating Velcro
pieces.
[0023] The receiver 145 provides the received TTL signal to a
microprocessor 150, which, in turn, controls electrodes that are
implanted in the animal's brain. A skull-top adapter 180 on the
animal houses the electrodes. A battery or other energizing means
may be housed in the backpack 160, which send electrical current to
the electrodes, e.g., energizes the electrodes, via short wires
170, to provide the desired stimulations to the brain sites to
which the electrodes are attached. In practice, a pair of wires and
electrodes is used for each brain site to be stimulated. Note that
the configuration shown is merely one possible example, which has
been found to be convenient for use by researchers. The particular
remote control set-up can be adapted to particular applications.
Moreover, additionally components may be carried by the backpack
160 or otherwise secured to the animal 190 including an upstream
transmitter for communication video data back to the operator.
[0024] Depending on the site of brain stimulation, an electric
stimulus can act as a cue or reward. Moreover, a reward stimulus
can act as a cue as well. While studies investigating such
phenomenon have generally been concerned with functional mechanisms
of the nervous system, little thought has been given to the
potential of behavioral paradigms constructed wholly around such
focal brain stimulations. Our study used stimulation of a reward
center of the brain to provide cues for moving forward, and
stimulation of portions of the brain that control left and right
movement as cues for moving left or right, respectively. For
example, the reward center may include the medial forebrain bundle
(MFB), ventral tegmental area, or other regions of the lateral
hypothalamus. The portion of the brain for controlling left and
right movement may include the somatosensory (SI) areas of the
brain, such as cortical representations of left and right whiskers
of the animal. In a particular experiment, SI and MFB stimulations,
which act as virtual cues and rewards, respectively, were delivered
to freely roaming rats. Behavioral contingencies were imposed so
that an operator could accurately steer the animal, in real-time,
over any arbitrarily specified 3-dimensional route and over any
real-world terrain.
[0025] We implanted stimulating electrodes in the MFB, plus right
and left SI whisker representations of a number of rats. The
whisker representations mimic the rat's sensation of being lightly
touched on the face. For example, if the rat has the sensation of
being touched on the right side of the face, e.g., as if the rat
was contacting a barrier on its right side, it will turn to the
left to avoid the barrier. Similarly, a sensation on the left side
of the face results in a right turn. The backpack 160, containing a
microprocessor-based remote-controlled microstimulator, was then
mounted on each animal. This allowed the operator using the laptop
computer 100 to directly deliver brief trains of 80 .mu.A stimulus
pulses to any of the implanted brain-sites at distances up to 500
meters (typically ten, 0.5 msec, biphasic pulses at 100 Hz).
Training the rats to navigate took ten sessions, during which the
animals learned to interpret remotely received brain stimulation as
instructions for directing their trajectory of locomotion. In a
figure-8 maze, they first learned to obtain periodic MFB rewards
(0.3-3 Hz) by running forward and turning correctly whenever left
or right turning cues were issued; these cues were presented as
"touch" stimulation of the left or right whiskers by stimulating
their respective cortical representations. The animals were then
placed in open environments that lacked the rectilinear structure
and fixed choice points of the maze.
[0026] All rats generalized their responses to their new
environments, running forward and turning instantaneously on cue
(see FIG. 2(a)). They moved at speeds averaging 0.3 m/s and worked
continuously for periods up to a 1-hour test limit. FIG. 2(a)
illustrates movement of the animal from a start point to a finish
point. The diagram was sketched from digitized video recordings.
Dark shaded dots indicate the rat's head positions at one-second
intervals. Light shaded dots indicate positions at which reward
stimulations were administered to the MFB. Light colored arrows,
labeled "R" or "L", indicate positions at which right or left
directional cues, respectively, were issued. Black arrows indicate
positions 0.5 sec after directional commands. Obstacles 205, 210,
215 and 220 created a slalom course. The inset indicates details of
the events that took place inside the dashed-line region.
[0027] Navigation over 3-D structures was achieved by incorporating
a unique behavioral attribute of MFB stimulation that reflected the
known "priming" qualities of MFB stimulation. We observed that MFB
stimulation not only reinforced forward locomotion but also
initiated and motivated further locomotion. Thus an MFB reward,
itself, served as an effective GO-forward cue. On approaching
objects such as a high step, GO-forward MFB stimulation would
induce the rats to climb or to descend from it. As a rule, the
number of such stimulations required was proportional to the
difficulty of the obstacle ahead (see FIG. 2(b)). The arrow and dot
depictions in FIG. 2(b) were obtained using the key as discussed in
connection with FIG. 2(a). Here, the rat 190 is guided over a
3-dimensional obstacle course. The animal was instructed to climb a
vertical ladder 250, cross a narrow ledge 260, descend a flight of
steps 270, pass through a hoop 280, and a descend a steep (70
degree) ramp 290. Two rounds of high-density MFB stimulation were
required to guide the rat successfully down the ramp, demonstrating
the motivational qualities of MFB-stimulation.
[0028] Superimposing GO-forward MFB stimulations onto the standard
schedule was thus sufficient to steer the rats through a wide
variety of complex, novel, and changing terrains. Our rats were
easily guided through pipes, across elevated runways and ledges,
and were induced to climb or jump from any surface that offered
sufficient purchase (e.g., trees). The animals were also guided to
systematically explore large collapsed concrete rubble piles and
directed through environments that they would normally avoid, such
as brightly lit wide-open outdoor arenas.
[0029] Our results show that "virtual" learning, by directly
accessing the central substrates of cues and rewards, can
effectively expand the scope of the operant method. It draws its
chief benefit from its ability to dissociate explicit schedule
variables such as cues and rewards from the physical variables
normally associated with their delivery, lending a freedom from the
mechanical and parametric constraints on learning imposed by
particular physical settings. The rewarding efficacy of MFB
stimulation is relatively non-satiating and animals need not
initiate consummatory behaviors to obtain them. Since virtual cues
and rewards are perceived within a body-centered frame of
reference, they may facilitate the learning of behaviors
independent of the external environment. It may also be possible to
increase the "bandwidth" conditionable information by stimulating
through a multiplicity of sites in the brain, thus increasing the
richness of elicited animal behaviors.
[0030] The specific behavioral model presented here--a guided
animal--has implications for new neurophysiological studies into
directed animal navigation. The model also represents a new
extension for operant conditioning into useful real-world
applications. Combined with electronic sensor and navigation
technology, the guided rat can be developed into an effective robot
platform possessing several natural advantages over current mobile
robots. Moreover, the added ability to remotely receive and
interpret brain activity allows the guided rat to function both as
mobile robot and biological sensor. This ability can be provided
using appropriate sensors and data transmitting equipment carried
by the rat.
[0031] FIG. 3 illustrates a schematic diagram of circuits for a
base station transmitter and a receiver. In one possible approach,
the animal's brain is stimulated using a multi-channel remote
control system that allows independent stimulation of each
electrode and therefore each associated brain region via each
channel. Note that the channels may be provided in various ways,
e.g., on separate carrier frequencies in a frequency division
multiplex, and/or in separate time slices in a time division
multiplex. Other various approaches for remotely controlling the
electrodes will be apparent to those skilled in the art. Moreover,
when multiple animals are controlled, appropriate techniques can be
used to ensure that only the intended animal is controlled, e.g.,
such as assigning frequencies or time slices to specific animals.
Spread spectrum communications may also be used, where spreading
and despreading codes are assigned for communications to specific
animals. The movements of the multiple animals may be coordinated
to achieve a specific goal. For example, if the animals are used to
provide surveillance of a building, each animal can be guided to a
specific, different part of the building. The cues can thus be
provided to the animals to coordinate their movements.
[0032] The invention provides a multi-channel telemetry brain
microstimulation system that provides a small, light, efficient and
reliable electrical stimulation platform, with flexibility for
experimental designs. The system includes two major components: the
base station transmitter 125 connecting through the serial port 115
to the PC 100, and a receiver-microprocessor 145, 150 integrated
into a backpack 160 that is carried by the animal 190, or otherwise
secured to the animal, e.g., using adhesive, or implanted in the
animal, such as under the animals skin.
[0033] In one possibility, the PC 100 issues ASCII stimulation
command strings, each containing an identifying header and desired
parameters. The microprocessor 150 translates the command, which is
relayed by the transmitter 125 and the receiver 145, into biphasic
TTL pulses to the specified channel. Oscilloscope recordings shows
that a stimulator associated with the receiver 145 executes the
stimulation commands with high fidelity and performs reliably even
in complex environments. A three-channel system was tested for
controlling three pairs of electrodes, one for stimulating the MFB,
and one each for the left and right side SI whisker stimulations.
However, this system is upgradeable to sixteen or more channels by
upgrading the microprocessor. The flexibility in programming
enables the system to deliver stimulation trains with different
parameters to different channels sequentially.
[0034] The use of multiple channels for controlling brain
stimulation allows multiple brain sites to be excited concurrently.
The present invention further provides high transmission fidelity,
reduced size and weight of the receiver that is implanted or
mounted on the animal, and reduced power requirements at the
transmitter and receiver. Moreover, the charge-balanced biphasic
pulses provided by the invention avoid electrolytic tissue injury
and electrode damage that can occur with monophasic pulses.
[0035] The system delivers brief trains of electrical stimulation
to three brain locations, each implanted with a pair of electrodes.
A program running on the laptop PC 100 specifies stimulation
parameters. The transmitter 125 sends out digital commands to the
receiver 145 and microprocessor 150 on the rat's backpack 160. The
microprocessor 150 executes the incoming command, resulting in an
output of a train of biphasic TTL pulses to the specified brain
location.
[0036] The PC program, written in BASIC, configures the serial port
115 to output stimulation-commands encoded by specific keystrokes
(keystrokes "j", "k" and "l" specified which of the three implanted
brain locations are stimulated). Instead of pressing keys on a
laptop computer, a dedicated controller with press buttons,
joystick or the like could be used. For each brain location, the
parameters of stimulation--the number of biphasic pulses in a
train, its frequency and the duration of each pulse--could be
specified. The commands were sent as an ASCII string, at 2400 baud,
from the PC to the transmitter 125 via the serial port 115. A short
header (e.g. "U", "U") was included to quiet inter-transmission
noise and to establish timing.
[0037] The transmitter circuit, shown generally at 300 in FIG. 3,
was built around a UHF transmitter (TX2, Radiometrix, Watford, UK)
powered by a 5V supply regulated from a 9-volt battery. First, the
serial port's RS232 signals were converted to TTL level signals
using an Agilent HCPL 2200 optocoupler and then sent to the TX2
transmitter 125. A quarter wavelength whip antenna 130 broadcast
the RF signal. The circuit 300 was put into an aluminum enclosure,
which served as circuit ground and RF ground plane. The TX2 module
125 is a two stage surface acoustic wave (SAW) controlled, FM
modulated transmitter that transmits at up to 40 kbps. It is
available in 433.93 MHz and 418 MHz versions, both of which we have
employed at the same time, with no cross talk.
[0038] The backpack circuitry, shown generally at 350 in FIG. 3,
was assembled on a printed circuit board. Its main components were
a receiver 145 (RX2, 5V version, Radiometrix) and a microprocessor
150 (Basic Stamp BS1 IC, Parallax Inc.) powered by a 6V 160-mAH
lithium battery (2CR-1/3N). The receiver used a helical antenna 140
(as described in the RX2 documentation). The backpack circuitry 350
included a light-emitting diode (LED) 355 that provided direct
visual verification of pulse delivery when the animals were freely
moving. The input/output (I/O) pins of the microprocessor were
connected to the skull-top adapter 180 that housed the electrode
ends by short flexible detachable cables. Under load (15 mA total),
without regulation, the 2CR-1/3N battery put out 5.5V. The
microprocessor ceased working when the battery voltage fell to
about 4.5V.
[0039] The Basic Stamp microprocessor 150 has eight tri-state
programmable digital I/O pins (P0-P7): one of these was set to
input the remotely received stimulation-command string and another
for output to the LED indicator. The remaining six were paired to
actuate three stimulus channels with biphasic pulse trains (thus
each channel used two I/O pins to stimulate its respective
electrode pair). The microprocessor was loaded with a PBASIC
program that controlled stimulation as follows: when not in use,
all electrode I/O pins were left in input mode
(Z.congruent.1M.OMEGA.) to prevent cross talk between electrodes.
For stimulation, a pair of pins were opened for output
(Z.congruent.20.OMEGA.) alternatively -5V applied first to one and
then the other. Since this system was floating, applying this
voltage to the first pin and then to the other resulted in a
biphasic pulse. After stimulation, the pins were reset to input
mode.
[0040] Under anesthesia, two teflon-coated stainless steel
microwire electrodes (100 .mu.m diameter), 1 mm apart, were
stereotaxically implanted in the MFB (left side) and the whisker
barrel fields (SI) of the two somatosensory cortices of five female
Long-Evans rats. Stimulation experiments commenced five days after
implantation.
[0041] FIG. 4 illustrates an oscilloscope trace of an ASCII command
string and the resulting train of biphasic pulses delivered to one
stimulation channel. The telestimulator followed remotely received
commands, which specified pulse frequency, pulse duration and the
number of pulses within a train, with high fidelity. The
microprocessor could deliver arbitrarily specified stimulus trains
from distances as much as 300 meters, line of sight. The 6 V,
160-mAh lithium backpack battery system survived about seven hours
of continuous stimulation (test stimulation consisted of stimulus
trains delivered at 0.2 Hz; each train each had five biphasic
pulses at 100 Hz with pulse duration 500 .mu.seconds). The
transmitter was able to work for several days (>7) using a 9 V
lithium battery. It weighed 268 g and could easily be carried,
along with the laptop, by the operator.
[0042] In particular, the ASCII stimulation command string (1) and
output stimulation waveform (2) taken from the oscilloscope (TDS
210, Tektronix). Channel 1 shows the TTL command on transmitter
input and channel 2 shows the biphasic stimulation waveform
(biphasic stimulus pulses) on one of the three channels of the
stimulator. It takes about 52 ms for the system to transmit and
execute the command. The stimulation waveform follows the specified
parameters: pulse number: 4, duration: 2.5 ms, and frequency: 100
Hz.
[0043] We next investigated the functional effectiveness of the
system. Specifically, our goal was to evaluate the behavioral
effectiveness of brain-stimulation delivered by direct 5V TTL
output of the microprocessor. This was done by observing
predictable behavioral responses consequent to stimulation of the
MFB. MFB stimulation is rewarding and can be used to condition
animal behaviors such as lever pressing. Connections were made
between the microprocessor outputs and the implanted electrodes,
and the rats were placed in a lever-equipped operant chamber in
which a train of biphasic pulses to the MFB followed each lever
press. Each train consisted of ten pulses delivered at 100 Hz with
pulse duration 0.7 msec. Under this reinforcement schedule all
subjects lever-pressed continuously, reaching pressing rates as
high as 150/min over a 20 min period. Using an oscilloscope hooked
across a resistor placed in series with the rat, we measured
electrode impedance (at 100 Hz) to be around 50-100 K.OMEGA.. Thus,
we estimated that the 5V TTL train delivered current amplitudes of
around 50-100 .mu.amps in the behaving animal.
[0044] We compared the functional efficacy of the constant voltage
5V source in generating MFB stimulation rewards to that of a
conventional constant current source which was set to deliver a
comparable pulse train at 100 .mu.amps (pulse duration 0.7 msec,
frequency 100 Hz, 10 pulses). FIG. 5 shows two averaged cumulative
frequency plots of the lever presses made by one rat over a
two-minute period using both the constant voltage (light shaded
line) and the constant current source (black line). The plots
obtained using these two techniques are similar in that they almost
overlap. We concluded that the telestimulator provided the
reliability and stationarity of stimulation behavioral effect equal
to that of a constant current stimulator set at 100 .mu.amps.
[0045] For field-testing, the backpack was mounted on the rats. The
animals were first trained to move forward continuously to obtain
periodic MFB stimulation. Thereafter, stimulation of SI cortex
(five pulses delivered at 100 Hz with pulse duration 0.5 msec)
served as directional cues, in that the animals learned to turn
left or right depending on which SI cortex was stimulated. Cortical
representations of left or right whiskers of the animal were
stimulated to effect a rightward or leftward change, respectively,
in the direction of movement. Under this basic reinforcement
contingency, we found that the rats could be accurately guided over
arbitrarily specified 3-D routes at considerable distances away,
showing that both cues and rewards were reliably delivered by the
telestimulation system. The rats worked without tiring for periods
up to a one-hour test limit.
[0046] The telestimulation system advantageously provides multiple
output channels in a single package to allow simultaneous bipolar
or monopolar stimulation of multiple brain sites, and it is both
reliable and robust as a brain-stimulator. A special feature of the
system is that it accomplished this task using conventional TTL
pulses. The use of tri-state logic circuitry to generate biphasic
stimulus pulses allows an experimenter to stimulate chronically
over long time periods while avoiding the electrolytic injury
caused by unidirectional currents. Another feature of the system is
that the backpack containing the receiver, microprocessor and
battery is small, light and power efficient, allowing it to be
carried by small animals over relatively long time periods. These
advantages are attributable to the relative simplicity of the
device, which uses a commercially available microprocessor to
provide well-controlled multichannel stimulus patterns. In our
case, the backpack microprocessor was programmed to carry out
specific stimulation patterns, but a simple reprogramming would
allow almost any pattern to be specified.
[0047] In our experiments, the telestimulation system was used to
generate behavioral effects in ways that make it possible to
develop new behavioral models for neurophysiology study in freely
moving animals. Somatosensory stimulation was used to create
percepts that were conditioned to act as cues in a behavioral task
reinforced by rewarding medial forebrain bundle stimulation. From
the point of view of generating sensory percepts and rewards the
fact that system output was a straightforward 5 V TTL pulse train
was not a limitation. The effect of a stimulus pulse train at any
given brain location depends on pulse amplitude, pulse duration,
pulse frequency, and the total number of pulses delivered. Within
certain windows these parameters are known to sum linearly. Thus,
changing one or more of the three programmable parameters (pulse
duration, frequency, and number of pulses) can create variable
desired magnitude of stimulation strength. In our study, these
stimulus parameters were arranged for optimizing the magnitude of
stimulus percepts and rewards (reward magnitude of MFB stimulation
was estimated by bar pressing rates in response to parametric
variation). As an example, FIG. 6 shows how bar pressing rates, in
one of our subjects, changed as a consequence of varying MFB
stimulation along a single dimension (data are from a single
session).
[0048] For the more general brain-stimulation experiment, however,
a constant voltage stimulator might be considered inadequate; some
experimental situations will require the ability to alter current
amplitude at will. If required, this additional capability--to make
it function as constant current source--can be also added onto our
basic telemetry system with some modification of circuitry. What
would be required includes !1) a higher voltage source, (2)
constant current circuitry, (3) a variable reference voltage for
constant current control of that circuitry (or a hardware set
current limit), and (4) an electrode isolation scheme, all small
enough to be incorporated into a reasonably sized backpack. As
noted above, since this mobile system is floating, it is not
necessary to have a split power supply for biphasic stimulation.
Requirement (1) could be met with charge pump devices. Requirement
(2) could be met with a power transistor in series with a current
sensing resistor and controlled by an operation amplifier.
Requirement (3) could be met with the Basic Stamp "PWM"
instruction, or, if larger microprocessors were to be used, a DAC.
Requirement (4) could be met with digital relays, some of which
have very high off impedances. One complete solution might be the
eight-channel application specific integrated circuit (ASIC) of
Troyk, though its current capabilities, as presently configured,
may not be adequate for MFB stimulation.
[0049] Moreover, the system tested could be modified in the
following ways: First, the number of bipolar stimulation channels
can be increased to sixteen by using a 32-channel microprocessor.
Increasing the baud rate of the serial communication system can
markedly reduce the stimulus delay. Alternatively, a look-up table
of commonly used stimulation patterns can be stored in the
microprocessor's memory, thereby reducing the number of characters
that must be sent through the serial port. Major advantages can be
obtained by integrating a general-purpose transceiver with a
microprocessor capable of handling multi-channel stimulation, and
also higher order I/O protocols. This would enable full duplex
wireless transmission enabling the delivery of stimulus commands to
the animal as well as receiving incoming sensor data if needed.
This wireless platform could serve to obtain neurophysiological
recordings from the behaving animal. Such a scheme, for example,
would be highly useful in studying the neural correlates of
navigation while directing freely moving animals.
[0050] Specific Applications for Remotely-Guided Animals
[0051] The techniques disclosed herein allowing human operators to
use wireless remote control to guide instrumented animals such as
rats through a range of terrestrial environments open up many
practical uses. In fact, such remotely guided animals (RGAs) may be
employed for important civilian, military and intelligence
applications, such as finding buried humans in collapsed buildings,
finding land mines, using stealth to gain intelligence about
explosives, drugs, or human targets in buildings, and placing
surveillance devices in trees, tunnels, and buildings.
[0052] The RGA fills a great need that exists for small robotic
devices that can carry a payload into areas that are dangerous,
inaccessible, or require stealth. By comparison with currently
available small robots, animals exhibit exceptional functionality
because evolution has imbued them with great adeptness in handling
different earthly terrains. Unlike existing artificial intelligence
devices which can be programmed to handle lists of pre-defined
tasks, biological intelligence has the innate ability to handle
novel real-world conditions. As such, RGAs are capable of great
autonomy in their ability to solve specific terrain problems. For
example, by simply rewarding animals such as rats for going right,
left or straight, by remote control stimulations to the brain, they
quickly learned to handle whatever obstacles they encountered.
Feral rats are particularly adept at penetrating buildings (e.g.,
by digging tunnels, squeezing through narrow cracks, chewing
through walls or jumping from trees), finding a target (e.g.,
food), and returning home. One of our overall aims here is to
harness such innate capabilities in animals that are also well
controlled. Moreover, RGAs carry their own highly developed natural
sensors that can be used for homing in on specific targets and for
detecting and adapting to novel situations. Rats' olfactory sense
is more acute, for example, than any artificial chemosensor.
Trained rats could replace dogs, therefore, as sensors for drugs or
explosives.
[0053] Most importantly, however, animals have excellent "sensor
fusion" capabilities, allowing them to combine different senses to
detect salient objects in the environment. A particular advantage
of rats is that they can use their non-visual senses to traverse
complex spaces in the dark. Thus, rats could be trained to find
humans trapped in collapsed buildings, for instance. Such humans
should be detectable not only by how they look, but also by their
scent, sound and feel. RGAs may also be useful for intelligence
operations because they are stealthy and naturally occurring. For
example, native feral rats could be implanted subcutaneously with
appropriate electronic devices, and then trained. Such rats could
then be returned to their environment, electronically guided to a
particular location, and used to transmit sensor information from
that location to a wireless receiving station.
[0054] For example, we have configured a system allowing video
signals to be returned from a rat to the operator. This system
includes a miniature CMOS video camera and a 2.4 GHz telemetry
transmitter, both of which are mounted on the animal's backpack.
The video camera is mounted on the animal's shoulders to provide a
"rat's eye" view of the local environment. The operator uses this
feedback to guide his decisions about the direction in which the
rat should be directed to turn. A microphone may also be used to
obtain audio data. Note that the animal may be guided in real-time
by a human operator or automatically under the control of a
computer that is programmed to issue appropriate instructions to
the animal. The animal may be guided automatically, analogous to an
autopilot system for an aircraft. If the animal veers off course or
some other anomalous condition is detected, an alarm notification
can be made to alert a human operator.
[0055] The backpack also includes a 9V Lithium battery pack, a
microprocessor with eight digital IO ports, and a 433 Mhz wireless
serial modem. Use of a serial modem to control an on-board
microprocessor enables a computer program to be downloaded into the
microprocessor, where the serial modem is then used to control its
application of stimulus trains through its different output ports.
When using low impedance electrodes, it is possible to deliver
stimuli directly through the microprocessor digital outputs
themselves. In general, the stimuli should be from 50-100
microamperes for rats. The stimuli parameters may vary for
different animals. The DIO has a 5V, 35 mA output capacity. This is
therefore sufficient if the electrodes have about 50 KOhm
resistance. Increasing pulse and train width, however, can
compensate for the low output voltage.
[0056] Possible applications are summarized below.
[0057] 1-Intelligence: Stealthily penetrate spaces (buildings,
encampments, installations). Explore and thereby map the space.
Gather specific information about activities in the space using
video cameras, microphones, and other sensors. Drop surveillance
devices in the space. Use remote communications (RF, ultrasound,
existing voice/data networks) to transmit information back to base.
Remain for long time periods in the space by foraging for food and
water. Harvest electrical power from the environment.
Alternatively, quickly return to base, carrying information stored
on backpack storage devices.
[0058] 2-Urban assault: Use rats to penetrate buildings and return
intelligence about enemy.
[0059] 3-Climb trees and other structures carrying equipment for
area surveillance.about.
[0060] 4-Penetrate tunnel networks: Rats enter tunnels and find
areas of human occupation
[0061] 5-Minefield: Remotely guide rat into suspected minefield.
Rat uses olfaction to sense buried mines. When mine chemicals
detected, the rat digs to find more chemical; This alerts the
operator about the location.
[0062] 6-Sentries: Rats guided to advance areas carry sensors,
and/or are sensors themselves for CBW agents.
[0063] 7-Drug or munitions sensing--cheaper than dogs. Can use
neurophysiological recording in olfactory system to identify
odorants.
[0064] 8-Search and rescue in rubble pile after building collapse.
(see below)
[0065] 9-Enter and return info about hard to enter small spaces in
buildings or under streets (HVAC, utility areas, cable conduits,
sewers). Find and possibly repair electrical, plumbing,
communications problems. Find and possibly remediate colonies of
insects, rodents, or other vermin.
[0066] 10-Climb trees to string Christmas lights.
[0067] Further details of possible applications are provided
below.
[0068] I. Search and Rescue in Collapsed Building.
[0069] A. A building falls because of structural failure, tornado,
earthquake, etc.
[0070] B. Rat is pre-trained to home on the scent, sound or sight
of human.
[0071] C. Rat is guided to the most likely site of entrance into
the rubble pile.
[0072] D. Rat caries electronics apparatus for remote guidance,
video with IR illumination, bi-directional audio, and accelerometer
and compass for spatial localization.
[0073] E. If human found, video can retrun picture. Through
bi-directional audio, have conversation with base.
[0074] F. If rat can drag a tube to the person, can supply water
and food through tube
[0075] G. If RF is lost:
[0076] Use building's residual electronic connections if
possible.
[0077] Rat drags tether with transceiver and water tube as far into
hole as possible.
[0078] Rat swarm is used, with additional rats guided to jettison
transceiver repeater units to relay signal to base.
[0079] Use other transmission medium, e.g., ultrasound.
[0080] Rat moves autonomously, and returns with info.
[0081] II. Search for Buried Landmines
[0082] A. Rats pre-trained to home in on scents of common landmine
chemicals.
[0083] B. Upon detecting the chemicals they begin to dig, and
continue as scent becomes stronger.
[0084] C. Rats are remotely guided through suspected landmine
fields.
[0085] D. Rats carry electronic apparatus for remote guidance,
video/audio, GPS, and accelerometer and compass for detection of
conditioned digging responses to finding a mine.
[0086] E. Rats are too small to set off mine, but can detect
them.
[0087] F. Accuracy of mine detection is enhanced by recording brain
activity characteristics of expected reward, or to scent-related
neural activity in olfactory lobe.
[0088] III. Reconnoiter Buildings in Terrorist or Urban Warfare
Situation
[0089] A. Rat is pre-trained to home in on humans while exploring
buildings.
[0090] B. Rat with standard instrumentation is remotely guided to a
suspect building.
[0091] C. Stealth achieved through darkness and use of appropriate
cover.
[0092] D. Rat enters building through holes in walls, windows,
doors, basement, roof, or sewer.
[0093] E. Rat explores building to find humans.
[0094] F. If RF is not lost, it returns video/audio and positions
of humans.
[0095] G. If RF is lost, it returns to base with pictures and
positions of humans.
[0096] Experimental Plan
[0097] A plan for enhancing RGA technology is provided to enhance
the current behavioral, neurophysiological, electronic,
communications and computer techniques used to implement RGAs for
specific applications. Particular emphasis is placed on the
following: 1-N expanding the range of semi-autonomous behaviors
that can be elicited though operator guidance, 2-improving the
technologies for teleoperation and realtime position sensing,
3-enhancing the sensor capabilities of the RGAs, both though
electronic and neurophysiological techniques, 4-developing tools
and techniques for coordinating swarms of RGAs, 5-improving the
cost and efficiency of RDAs by developing automated training
techniques, and 6-improving the RGAs' stealth and longevity by
developing subcutaneous implantable electronics for communications,
control and energy harvesting. Various factors are provided
below.
[0098] I Animal
[0099] A. Species
[0100] 1. Rat--public relations, stealth, versatility, indoor
terrains
[0101] 2. Squirrels--arboreal adeptness, stealth
[0102] 3. Rabbits--speed, stealth
[0103] 4. Coyotes, jackals--public relations, stealth,
adeptness
[0104] 5. Larger mammals--payload, specific capabilities
[0105] 6. Birds--flight, soaring, stealth
[0106] B. Gender
[0107] 1. Females--accurate guidance, terrain adeptness
[0108] 2. Males--payload, endurance
[0109] C. Size
[0110] 1. Max payload about {fraction (1/3)} of body weight
[0111] 2. Genetically large strains
[0112] 3. Growth hormone enhanced
[0113] D. 300 g hooded rat female--best for training
[0114] E. 700 g hooded rat male--best for payload
[0115] F. Large strains of rats? Growth hormone?
[0116] G. 700 g feral rat--best for terrain capability and
stealth
[0117] H. Other mammal
[0118] I. Bird
[0119] II. Behavioral
[0120] A. Specific movements
[0121] 1. Left/Right
[0122] 2. Stop
[0123] 3. Rear
[0124] 4. Move head to position mini camera
[0125] B. Finding specific targets
[0126] 1. Olfactory
[0127] 2. Auditory
[0128] 3. Type of space, e.g. Open doorway
[0129] 4. Living creature, e.g. human
[0130] C. Autonomies
[0131] 1. Explore a space, i.e. follow a wall around a building
[0132] 2. Go to a buil ding
[0133] 3. Go fast straight ahead
[0134] 4. Go through a tunnel
[0135] 5. Go up a staircase
[0136] D. Terrains
[0137] 1. Indoor rooms, hallways
[0138] 2. Indoor industrial
[0139] 3. Inside walls
[0140] 4. Under streets, sewers
[0141] E. Learning, training techniques
[0142] 1. Train sequentially, or randomly
[0143] 2. Training to handle distractions
[0144] 3. Training for autonomous functions
[0145] 4. Training for foraging and handling environments
[0146] 5. Training for stealth
[0147] F. Automated training
[0148] 1. Accelerometer to sense turning
[0149] 2. Reward turning with appropriate timing
[0150] III. Neurophysiological
[0151] A. What is best way to do Left/Right command?
[0152] 1. Slstim
[0153] 2. Mlstim
[0154] 3. PPstim
[0155] 4. Piezoelectric sound
[0156] B. Best way to do higher level commands
[0157] 1. Sensory cortical areas
[0158] 2. Brain areas matched to command function, i.e. Hypothal
for stop.
[0159] 3. Use intrinsic brain maps to grade command;
whisker--direction
[0160] C. Neurophysiological recording for sensor
[0161] 1. Olfactory bulb--smell
[0162] 2. Hippocampus--spatial position
[0163] 3. Sornatosensory vibrissa--aperture shape
[0164] D. Neurophysiological recording for higher functions
[0165] 1. VTA, accumbens
[0166] 2. Motor command--to control external device
[0167] 3. Intention--to communicate with swarm
[0168] IV. Neurophysiology Technology
[0169] A. Electrodes
[0170] 1. Up to 16 bipolar channels
[0171] 2. Spaced multicontact arrays
[0172] 3. Platinum or gold contacts
[0173] 4. Steroid eluting to prevent bioreaction
[0174] B. Multichannel stimulator
[0175] 1. 16 bipolar channels
[0176] 2. Handles bipolar or monopolar
[0177] 3. Bipolar pulse sequences
[0178] 4. Isolated
[0179] 5. Programmable current
[0180] 6. Programmable pulse trains
[0181] 7. Simultaneous stimulations, interlaced
[0182] 8. Miniaturized, low power
[0183] 9. Subdermally implantable
[0184] 10. Hard wired to electrodes for subdermal implantation
[0185] C. Micro controller
[0186] 1. Handles full duplex signal transmission from operator or
swarm
[0187] 2. Controls stimulator
[0188] 3. Handles acceleration, position sensing devices
[0189] 4. Controls video frame transmission
[0190] 5. Controls jettison of sensor devices
[0191] 6. Controls data storage in local memory buffer
[0192] 7. Handles autonomous functions
[0193] 8. Node for swarm coordination
[0194] 9. Repeater for swarm signals
[0195] D. Wireless data transceiver
[0196] 1. Wireless full duplex serial interface to
microcontroller
[0197] 2. Intermittent 56K baud
[0198] 3. Spread spectrum
[0199] Beyond the use of RGAs to perform missions, they also
provide valuable research models for the development of biomimetic
robots. Engineers wishing to understand how animals handle
difficult terrain and solve complex problems have already obtained
important insights from biomechanical and neurophysiological
analyses of animals. RGAs are advantageous in that they can be
directed to perform precise experimental procedures. Such tightly
controlled experiments will be necessary to truly understand the
biological mechanisms underlying animals' extraordinary abilities
to handle real world problems.
[0200] Moreover, the brains of mammals and birds all possess
mesolimbic dopamine fibers in their lateral hypothalamic medial
forebrain bundles. This system is considered a final common pathway
for reward and motivation. Mild electrical stimulation of this
region in birds, as well as mammals can mimic the rewarding effects
of physical reinforcers, such as food or water. Thus, different
mammalian or avian species could be used for different RGA
applications. For example, a remotely guided urban pigeon carrying
a small video camera could obtain a wealth of information.
[0201] Finally, the use of animals for such purposes is quite
humane. Beyond the discomfort associated with recovery from the
implant surgery, these animals suffer no pain, and need not be
sacrificed. Since the training involves rewards, it is superior to
punishment based training methods. Moreover, the use of animals to
serve a human need is consistent with the long-established human
history of domesticating animal species based on their evolutionary
adaptations. Laboratory rodents, which are new symbionts with
humans, were domesticated because they are inexpensive, robust and
arouse relatively little sentimental attachment in humans. The same
qualities make them good candidates for RGA applications.
[0202] The invention has been described herein with reference to
particular exemplary embodiments. Certain alterations and
modifications may be apparent to those skilled in the art, without
departing from the scope of the invention. The exemplary
embodiments are meant to be illustrative, not limiting of the scope
of the invention, which is defined by the appended claims.
* * * * *